Learning deep latent space for unsupervised violence detection

Ehsan, Tahereh Zarrat; Nahvi, Manoochehr; Mohtavipour, Seyed Mehdi

doi:10.1007/s11042-022-13827-7

Cited by 5 publications

(4 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…According to our survey, the lack of a public dataset for chicken welfare analysis in poultry houses presents a significant challenge in this field, limiting researchers' ability to work on this topic. Additionally, the number of research papers on chicken behavior analysis lags considerably behind other computer vision domain such as human behavior analysis [22][23][24] and human abnormal behavior detection [25][26][27].…”

Section: Materials and Methods 21 Datasetmentioning

confidence: 99%

DABA-Net: Deep Acceleration-Based AutoEncoder Network for Violence Detection in Surveillance Cameras

Ehsan

Nahvi

Mohtavipour

2022

2022 International Conference on Machine Vision and Image Processing (MVIP)

View full text Add to dashboard Cite

Detecting anomalies in poultry houses is crucial for maintaining optimal chicken health conditions, minimizing economic losses and bolstering profitability. This paper presents a novel real-time framework for analyzing chicken behavior in cage-free poultry houses to detect abnormal behaviors. Specifically, two significant abnormalities, namely inactive broiler and huddling behavior, are investigated in this study. The proposed framework comprises three key steps: (1) chicken detection utilizing a state-of-the-art deep learning model, (2) tracking individual chickens across consecutive frames with a fast tracker module, and (3) detecting abnormal behaviors within the video stream. Experimental studies are conducted to evaluate the efficacy of the proposed algorithm in accurately assessing chicken behavior. The results illustrate that our framework provides a precise and efficient solution for real-time anomaly detection, facilitating timely interventions to maintain chicken health and enhance overall productivity on poultry farms.

show abstract

Section: Materials and Methods 21 Datasetmentioning

confidence: 99%

DABA-Net: Deep Acceleration-Based AutoEncoder Network for Violence Detection in Surveillance Cameras

Ehsan

Nahvi

Mohtavipour

2022

2022 International Conference on Machine Vision and Image Processing (MVIP)

View full text Add to dashboard Cite

show abstract

“…While studying all these approaches based on spatial and/or temporal features, several other models like Efficient 3D CNN [37], Xception + BiLSTM + Attentions [38], C3D [39], AlexNet + LSTM [40], Hough Forests + 2D CNN [41], Three Streams + LSTM [42], MoSIFT [43], motion intensities + AdaBoost [44], ResNet50 + ConvLSTM [45], Fine‐tuned MobileNet [46] and Motion Blobs + Random Forest [47] and Double‐AE [48] have also attempted to address the violence detection problems and demonstrated notable results. Hence, to better understand the progress of AVDC so far, the prominent approaches discussed above have been summarized in Table 1.…”

Section: Related Workmentioning

confidence: 99%

BrutNet: A novel approach for violence detection and classification using DCNN with GRU

Haque,

Nyeem,

Afsha

2024

The Journal of Engineering

View full text Add to dashboard Cite

Automatic Violence Detection and Classification (AVDC) with deep learning has garnered significant attention in computer vision research. This paper presents a novel approach for combining a custom Deep Convolutional Neural Network (DCNN) with a Gated Recurrent Unit (GRU) in developing a new AVDC model called BrutNet. Specifically, a time‐distributed DCNN (TD‐DCNN) is developed to generate a compact 2D representation with 512 spatial features per frame from a set of equally‐spaced frames of dimension 16090 in short video segments. Further to leverage the temporal information, a GRU layer is utilised, generating a condensed 1D vector that enables binary classification of violent or non‐violent content through multiple dense layers. Overfitting is addressed by incorporating dropout layers with a rate of 0.5, while the hidden and output layers employ rectified linear unit (ReLU) and sigmoid activations, respectively. The model is trained on the NVIDIA Tesla K80 GPU through Google Colab, demonstrating superior performance compared to existing models across various video datasets, including hockey fights, movie fights, AVD, and RWF‐2000. Notably, the model stands out by requiring only 3.416 million parameters and achieving impressive test accuracies of 97.62%, 100%, 97.22%, and 86.43% on the respective datasets. Thus, BrutNet exhibits the potential to emerge as a highly efficient and robust AVDC model in support of greater public safety, content moderation and censorship, computer‐aided investigations, and law enforcement.

show abstract

“…Recently, Vijeikis et al 34 time-distributed MobileNetV2 and LSTM-based model for addressing an efficient violence detection problem. For higher classification accuracy, Ehasan et al 35 proposed an UNet + PatchGAN-based unsupervised action translation network utilizing spatio-temporal features to identify violent behaviours and overcome the problem related to the insufficiency of relevant data. Similarly, Mohtavipour et al 36 proposed a multi-stream CNN-based AVDC approach.…”

Section: Spatio-temporal Feature-based Modelsmentioning

confidence: 99%

“…While studying all these approaches based on spatial and/or temporal features, several other models like Efficient 3D CNN 37 , Xception + BiLSTM + Attentions 38 , C3D 39 , AlexNet + LSTM 40 , Hough Forests + 2D CNN 41 , Three Streams + LSTM 42 , MoSIFT 43 , motion intensities + AdaBoost 44 , ResNet50 + ConvLSTM 45 , Fine-tuned MobileNet 46 and Motion Blobs + Random Forest 47 and Double-AE 48 have also attempted to address the violence detection problems and demonstrated notable results. Hence, to better understand the progress of AVDC so far, the prominent approaches discussed above have been summarized in the Table . 1.…”

Section: Spatio-temporal Feature-based Modelsmentioning

confidence: 99%

BrutNet: A Novel Approach for Violence Detection and Classification using DCNN with GRU

Haque,

Nyeem,

Afsha

2024

Preprint

View full text Add to dashboard Cite

Automatic Violence Detection and Classification (AVDC) with deep learning has garnered significant attention in computer vision research. This paper presents a novel approach to combining a custom Deep Convolutional Neural Network (DCNN) with a Gated Recurrent Unit (GRU) in developing a new AVDC model called BrutNet. Specifically, we develop a time-distributed DCNN (TD-DCNN) to generate a compact 2D representation with 512 spatial features per frame from a set of equally-spaced frames of dimension 160×90 in short video segments. Further to leverage the temporal information, a GRU layer is utilised, generating a condensed 1D vector that enables binary classification of violent or non-violent content through multiple dense layers. Overfitting is addressed by incorporating dropout layers with a rate of 0.5, while the hidden and output layers employ rectified linear unit (ReLU) and sigmoid activations, respectively. The model is trained on the NVIDIA Tesla K80 GPU through Google Colab, demonstrating superior performance compared to existing models across various video datasets, including hockey fights, movie fights, AVD, and RWF-2000. Notably, our model stands out by requiring only 3.416 million parameters and achieving impressive test accuracies of 97.62%, 100%, 97.22%, and 86.43% on the respective datasets. Thus, BrutNet exhibits the potential to emerge as a highly efficient and robust AVDC model in support of greater public safety, content moderation and censorship, computer-aided investigations, and law enforcement.

show abstract

Learning deep latent space for unsupervised violence detection

Cited by 5 publications

References 40 publications

DABA-Net: Deep Acceleration-Based AutoEncoder Network for Violence Detection in Surveillance Cameras

DABA-Net: Deep Acceleration-Based AutoEncoder Network for Violence Detection in Surveillance Cameras

BrutNet: A novel approach for violence detection and classification using DCNN with GRU

BrutNet: A Novel Approach for Violence Detection and Classification using DCNN with GRU

Contact Info

Product

Resources

About